Combating Antimicrobial Resistance with Extended Beta ... · PDF fileObjectives • Review pharmacokinetic and pharmacodynamic (PK/PD) properties of beta-lactam antibiotics • Evaluate

Combating Antimicrobial Resistance with Extended Infusion Beta-lactamsStephen Andrews, PharmDPGY-1 Pharmacy Practice Resident

Disclosure• The presenter has no conflicts of interest to

disclose with material in this presentation.

4/3/2017 2

Objectives• Review pharmacokinetic and pharmacodynamic

(PK/PD) properties of beta-lactam antibiotics

• Evaluate the literature for using extended infusion beta-lactam antibiotics

• Determine candidates for extended infusion beta-lactam therapy

• Identify obstacles in administering extended infusion beta-lactam therapy

4/3/2017 3

No Longer a Threat: Reality of Resistance

• “Pan-Resistant New Delhi Metallo-Beta-Lactamase-Producing Klebsiella pneumoniae” -Washoe County, Nevada– Resistant to 26 antibiotics

• 10-50% of patients treated for nosocomial pneumonia develop antimicrobial resistance

• No new classes of antibiotics since the 1980s

4/3/2017 4Asin-Prieto E, et al. J Infect Chemother 2015;21:319-329Bao H, et al. Eur J Clin Microbiol Infect Dis. 2017;36(3):459-466MMWR Morb Mortal Wkly Rep 2017:66:33

Isolate Susceptibility – Antibiogram 2015Piperacillin-tazobactam

Cefepime Ertapenem Meropenem

P. aeruginosa(n=1,065)

66% 77% -- 66%

E. coli(n=3,481)

96% 91% 100% 100%

K. pneumoniae(n=881)

95% 99% 100% --

A. baumanii(n=119)

42% 69% -- 79%

4/3/2017 5The University of Kansas Health System, Clinical Lab

Outline• Common PK/PD parameters

• Monte Carlo simulations

• Literature evaluation

• Practical considerations

4/3/2017 6

Time-Dependent (Time > MIC)

• Longer duration of exposure above the MIC results in greater bactericidal activity– Beta-lactam antibiotics

4/3/2017 7MIC: Minimum Inhibitory ConcentrationCraig WA. Clin Infect Dis. 1998 Jan;26(1):1-10

Concentration-Dependent (Peak / MIC)

• Higher concentrations result in greater bactericidal activity– Aminoglycosides

4/3/2017 8MIC: Minimal inhibitory concentrationCraig WA. Clin Infect Dis. 1998 Jan;26(1):1-10

AUC-Dependent (AUC/MIC)

• Larger AUCs above the MIC result in greater extent of killing– Fluoroquinolones

4/3/2017 9Craig WA. Clin Infect Dis. 1998 Jan;26(1):1-10

Beta-lactams: Time-Dependent• Approved dosing regimens occurred before robust

PK / PD studies

• Bactericidal response is predicted by the amount of time above the MIC

• Regrowth occurs rapidly when concentrations are below MIC between administrations– Chance for resistance increases

4/3/2017 10Osthoff M, et al. Swiss Med Wkly. 2016;146:w14368Craig WA. Clin Infect Dis. 1998 Jan;26(1):1-10

Options for Increasing Time above MIC

• Option 1: Increase the dose– Increases risk of concentration-dependent toxicities

• Option 2: Extend the infusion time

4/3/2017 11Craig WA. Clin Infect Dis. 1998 Jan;26(1):1-10

Duration Above MIC is Drug-Specific

• Optimal time above MIC by drug class– Carbapenems: 30-60%– Cephalosporins: 60-70%– Penicillins: 50-60%

• PK / PD parameters initially studied in neutropenic animal models

• Validated in healthy adults

4/3/2017 12Craig WA. Clin Infect Dis. 1998 Jan;26(1):1-10Drusano G. Clin Infect Dis. 2003;36:S42-50Drusano G. Nat Rev Microbiol. 2004;2:289-300

Pharmacodynamic Changes in Illness

4/3/2017 13Osthoff M, et al. Swiss Med Wkly. 2016;146:w14368

Critical Illness

Inflammation Hypotension

Volume Resuscitation

Increased Volume of

Distribution

Acute Kidney Injury

Decreased Clearance

Increased Cardiac Output

Increased Clearance

Infusion Duration affects Time > MIC


Therapeutic Drug Monitoring• Not routine practice for beta-lactams

• Monte Carlo simulations used to determine optimal dosing strategies

• Statistical model to predict achievement of a PK / PD target– Probability of target attainment– Cumulative fraction of response

4/3/2017 15Asin-Prieto E, et al. J Infect Chemother 2015;21:319-29Mouton JW, et al. J Antimicrob Chemother 2005;55:601-7

Review Question 1• Which of the following describes the PK / PD

parameter for beta-lactam antibiotics?

A. Concentration / MIC

B. Time / MIC

C. AUC / MIC

4/3/2017 16

What is the clinical evidence supporting extended infusion beta-lactam antibiotics?

4/3/2017 17

Extended Infusion vs. Traditional Dosing Piperacillin-tazobactam with P. aeruginosa

Objective

• Evaluate efficacy of extended infusion piperacillin-tazobactam vs. traditional dosing for P. aeruginosa infections

Outcomes

• 14-day Mortality• Length of Hospital Stay

Dosing

• Extended Infusion: 3.375g q8h over 4 hours• Traditional: 3.375g q6h over 30 minutes

4/3/2017 18EI: Extended InfusionLodise TP, et al. CID 2007;44:357-63

Extended Infusion vs. Traditional Dosing Piperacillin-tazobactam with P. aeruginosa

Inclusion Criteria

• Adults with confirmed P. aeruginosa infections• Absolute neutrophil count >1000• Piperacillin-tazobactam given within 72 hours of

infection and continued at least 48 hours• ICU and floor patients

Exclusion Criteria

• > 1 day of traditional dosing before extended infusion• Reported resistance• Dialysis, cystic fibrosis, solid organ or bone marrow

transplant patients • Beta-lactam therapy within 5 days of initiation of

piperacillin-tazobactam

4/3/2017 19Lodise TP, et al. CID 2007;44:357-63

Study Design• Single-center, retrospective cohort study

• No loading dose reported– Loading dose: same dose given as an intermittent

infusion prior to beginning extended infusion


Justification of Dosing Strategy• High probability of achieving piperacillin

concentrations above MIC for 50% of dosing interval (50% ƒT>MIC)


Patient Characteristics

Source of Infection Extended(n = 102)

Intermittent(n = 92) P - value

Pulmonary (%) 55 (53.9) 48 (52.2) 0.8

Intra-abdominal (%) 4 (3.9) 1 (1.1) 0.2

Urinary (%) 21 (20.6) 12 (13) 0.2

Skin & soft tissue (%) 11 (10.8) 23 (25) 0.009

Bloodstream (%) 3 (2.9) 0 (0) 0.1

Other (%) 8 (7.8) 8 (8.7) 0.8


• 126 patients (64.9%) in the ICU when therapy started

Clinical Outcomes• Classification and Regression Tree (CART)

analysis– Goal: Find the breakpoint APACHE II score for benefit

• Extended infusion dosing decreased mortality and length of stay (LOS) if APACHE II ≥ 17– Average APACHE II score = 15.6

4/3/2017 23

Endpoint Extended(n=102)

Traditional(n=92) p-value

Mortality (%) 12.2 (5 / 41) 31.6 (12 / 38) 0.04Hospital LOS (days) 21 (3 - 91) 38 (6 - 131) 0.02

Lodise TP, et al. CID 2007;44:357-63

Limitations• Retrospective study design

• Potentially suboptimal dose for P. aeruginosa in traditional infusion group

• Only P. aeruginosa confirmed infections

• No MIC data available

• Variable indications for treatment

• No subgroup analysis based on infection source


Unanswered Questions• Does the benefit of extended infusion apply to all

gram-negative pathogens or only Pseudomonas ?

• Should the MIC dictate whether extended infusions are necessary?

• Do extended infusion beta lactam dosing strategies improve mortality outcomes beyond 14 days?

4/3/2017 25

Extended Infusion vs. Traditional Dosing with Gram-negative Infections

Objective

• Evaluate efficacy of extended infusion piperacillin-tazobactam vs. traditional dosing for gram-negative infections

Outcomes

• 30-day Mortality• Length of Hospital Stay• Results stratified by MIC: <8, 8 to 16, >16 mg/L

Dosing

• Extended Infusion: 3.375g q8h over 4 hours• Traditional: 3.375g to 4.5g q6-8h over 30 minutes

4/3/2017 26Patel GW, et al. Diagnostic Microbiology and Infectious Disease 64 (2009) 236-240

Extended Infusion vs. Traditional Dosing with Gram-negative Infections

Inclusion Criteria

• Age 18 or older, ANC ≥1000• Confirmed infection with gram-negative pathogen• Piperacillin-tazobactam given within 72 hours of

infection continued at least 48 hours

Exclusion Criteria

• >1 day of traditional piperacillin-tazobactam before EI• Reported resistance• Dialysis, cystic fibrosis, solid organ or bone marrow

transplant patients• Concurrent beta-lactam therapy within 5 days of

initiation of piperacillin-tazobactam


Study Design• Retrospective, multisite cohort study

• Power not set


Patient Characteristics

4/3/2017 29

Source of Infection Extended(n=70)

Intermittent(n=59)

Pulmonary (%) 17 (20.4) 20 (33.9)Intra-abdominal (%) 9 (12.9) 10 (16.9)Urinary (%) 27 (38.6) 18 (30.5)Skin & soft tissue infection (%) 10 (14.3) 10 (16.9)Bloodstream (%) 4 (5.7) 1 (1.7)Other (%) 3 (4.3) 0(0)

Organisms Present(>10% frequency)

Extended(n=70)

Intermittent(n=59)

Escherichia coli 27 (28.6) 23 (39)Psuedomonas aeruginosa 14 (20) 14 (23.7)Klebsiella pneumoniae 14 (20) 12 (20.3)Proteus spp. 3 (4.3) 13 (22)

Patel GW, et al. Diagnostic Microbiology and Infectious Disease 64 (2009) 236-240

Clinical Outcomes• No statistical difference in 30 day mortality or

length of hospital stay

4/3/2017 30


Intermittent(n=59)

Mortality (%) 5.7 8.5

Hospital LOS (days) - Overall 8 (5.5-15) 8 (5-11)Hospital LOS (days) – MIC <8 mg/L(n=76) 8 (5.5-15) 8 (5-11)

Hospital LOS (days) – MIC 8-16 mg/L(n=52) 5 (4-10.5) 5 (4-9)

Hospital LOS (days) – MIC >16 mg/L(n=1) N/A 17 (17)

APACHE II Score, mean (SD) 10.9 (5.3) 10.5 (5.5)

Patel GW, et al. Diagnostic Microbiology and Infectious Disease 64 (2009) 236-240

Why did the study fail to find benefit in using extended infusions?• Smaller study size (n=129)

• 43% of patients in traditional dosing group had a CrCl <40 mL/min

• 59% of patients in traditional dosing had MIC <8 mg/L

• 35% of patients had a urinary infection


Summary• More likely to benefit:

– Pseudomonal strains– Pulmonary source of infection– APACHE II ≥ 17

• Less likely to benefit:– Gram-negative infections with lower MICs– Patients with urinary tract infections

• Does this phenomenon hold up when compared to all beta-lactams with similar spectrums?

4/3/2017 32

Extended Infusion vs. Traditional Dosed Beta-lactams with Gram-negative Infections

Objective

• Compare clinical outcomes for extended infusion piperacillin-tazobactam versus traditional dosing of similar spectrum beta-lactams in gram-negative infections

Outcomes

• In-hospital mortality rate• Length of hospital and ICU stay• Total duration of antibiotic therapy

Dosing

• Extended-infusion piperacillin-tazobactam: 3.375g q8h over 4 hours

• Traditional Dosing: cefepime, ceftazidime, imipenem-cilastatin, meropenem, doripenem, or piperacillin-tazobactam

4/3/2017 33Yost, RJ, et al. Pharmacotherapy 2011;31(8):767-775

Extended Infusion vs. Traditional Dosed Beta-lactams with Gram-negative Infections

Inclusion Criteria

• Adults with confirmed gram-negative infections• Mixed infections were allowed

• Patients hospitalized for >72 hours• Antibiotics given for >48 hours

Exclusion Criteria

• >1 day of traditional piperacillin-tazobactam before extended infusion

• Intermediate or resistant to initial therapy• Concomitant beta-lactam therapy• Gram-positive or fungal coverage was inappropriate


Study Design

• Multicenter, retrospective study

• Included a multivariate analysis

4/3/2017 35

Clinical Outcomes

• Extended infusion piperacillin-tazobactamreduced in-hospital mortality

• No difference in antibiotic duration, hospital or ICU length of stay

4/3/2017 36


Intermittent(n=173) p-value

Mortality (%) 9.7% 17.9% 0.02

Yost, RJ, et al. Pharmacotherapy 2011;31(8):767-775

Multivariate Endpoint Odds Ratio 95% CI p-value

Mortality 0.22 0.07-0.76 0.053

Survival (days) 2.77 0.85-4.7 <0.01

Source of Infection

4/3/2017 37

Source of Infection Extended(n=186)


Pulmonary (%) 57 (30.7) 75 (43.3) 0.01

Urinary (%) 76 (40.9) 63 (36.4) 0.39

Skin & soft tissue infection (%) 36 (19.4) 35 (20.2) 0.82

Bloodstream (%) 41 (22) 47 (27.2) 0.26

Other (%) 13 (7) 28 (16.2) 0.01


Potential Confounders

4/3/2017 38

Group characteristics Extended(n=186)


Pseudomonas sp. 42 (22.6%) 69 (39.9%) <0.01

Streptococcus sp. 3 (1.6%) 13 (7.5%) 0.01

IV aminoglycosides 11 (5.9%) 28 (16.2) <0.01

APACHE II -- -- 0.07


Does this show extended infusion beta-lactam dosing is superior?Supportive

• Extended infusion outperformed combination therapy with aminoglycosides (5.9% vs. 16.2%)

Confounding

• More P. aeruginosa in traditional dosing group

– Potential higher MIC

• More respiratory tract infections with traditional dosing group

– Potentially more difficult to treat

– No group specific mortality information published


Review Question 2• Which of the following groups are more likely to

benefit from extended infusion beta-lactam therapies?

A. Patients who are more critically ill

B. Patients who have P. aeruginosa infections

C. Patients who have urinary tract infections

D. Both A and B

4/3/2017 40

Empiric Coverage for SuspectedGram-negative Infections

Objective

• Determine if extended infusion beta-lactams improve outcomes in critically ill patients with suspected or confirmed gram-negative infections• Expect lower rate of mortality based on patient population

Primary Outcome

• Resolution of fever to < 38.3°C and > 36°C for 24 consecutive hours, and/or

• WBC decrease to < 12,000/μL or 50% decrease within 7 days

Secondary Outcomes

• Time to defervescence• All-cause hospital-, 14-, and 30-day mortality• ICU and hospital length of stay

4/3/2017 41Arnold HM, et al. Ann Pharmacother 2013;47:170-80

Standardized Dosing Schemes


• All extended infusions occurred over 3 hours

Drug CrCl (ml/min) IV Dose

Cefepime ≥6030-60

2g every 8 hours2g every 12 hours

Meropenem ≥5030-50

1g every 8 hours1g every 12 hours

Piperacillin-tazobactam ≥4030-40

4.5g every 6 hours3.375g every 6 hours

Empiric Coverage for Suspected Gram-negative Infections

Inclusion Criteria

• ICU status receiving empiric gram-negative coverage• Suspected health-care associated infection• Fever (>38.3°C) and/or WBC >12,000/μL• Cultures drawn within 24 hours of the antibiotic start

time

Exclusion Criteria

• <48 consecutive hours of antibiotics while in ICU• CrCl <30 ml/min or renal replacement therapy• Antibiotic dose outside protocol• Urinary tract infection• Only gram-positive, fungal, or viral pathogen identified


Study Design• Single center, pre/post implementation trial

• 162 patients needed to detect 50% relative risk reduction in treatment failure


Patient Characteristics• 503 patients included


Group characteristics

Extended(n=261)


COPD 57 (21.8%) 81 (33.5%) 0.003Cefepime 115 (47.5%) 143 (54.8%) 0.103Meropenem 86 (35.5%) 64 (24.5%) 0.007Piperacillin-tazobactam 41 (16.9%) 54 (20.7%) 0.283

APACHE score,median (IQR) 21 (16-25) 19 (17-24) 0.466

Clinical Outcomes• No statistically significant difference in primary

outcome or secondary outcomes


Endpoint Extended (n=261)


Resolution of fever and WBC 51% 56.6% 0.204

30 day Mortality (%) 25.7% 23.6% 0.542

ICU LOS (days) 10.8 (5-17) 9.3 (5.6-19.2) 0.138

Limitations• Only 41% of patients had confirmed infections

• Limited number of patients with high MIC organisms identified:– n=25 of 206 patients with identified organisms– Enterobacteriaceae, P. aeruginosa, Acinetobacter

• No loading dose given– Low serum concentrations initially could impact

outcomes– Time to first antibiotic doses are determinants of

mortality

4/3/2017 47Arnold HM, et al. Ann Pharmacother 2013;47:170-80Gaieski DF, et al. Crit Care Med 2010;38(4):1045-53

Does prospective data show similar results?

4/3/2017 48

Continuous vs. Intermittent Infusions BLING II & BLISSStudy BLING II (2015) BLISS (2016)

Objective Evaluate efficacy of continuous infusion beta-lactams in critically ill patients with severe sepsis

Primary Outcome

Number of alive ICU-freedays at Day 28

Clinical cure at 14 days after antibiotic cessation

Secondary Outcome(s)

- 90-day mortality- Clinical cure at 14 days

after antibiotic cessation- Alive organ failure-free

days at Day 14- Duration of bacteremia

post randomization

- 14- and 30-day survival- PK/PD target attainment- ICU and ventilator-free

days at 28 days post randomization

4/3/2017 49Abdul-Aziz, MH, et al. Intensive Care Med 2016;42:1535-1545Dulhunty JM, et al. Am J Respir Crit Care Med 2015;192(11):1298-1305


Study Design - Randomized controlled trial

- Open label

- Randomized controlled trial

- Double-blinded

Medications Similar spectrum of activity

- Piperacillin/tazobactam- Ticarcillin/clavulanate- Meropenem

- Piperacillin/tazobactam- Cefepime- Meropenem

Dosing Loading dose givenAllowed concomitant, non-beta-lactam antibiotics



Inclusion Criteria

≥18 years oldSevere sepsis criteria

Exclusion Criteria

- >24 hours beta-lactam before randomization

- Pregnancy- Allergy to study drug

- Continuous renal replacement therapy

- Impaired hepatic function- Inadequate central venous

access


Patient CharacteristicsBLING II

4/3/2017 52Dulhunty JM, et al. Am J Respir Crit Care Med 2015;192(11):1298-1305

Group characteristics Continuous(n=212)

Intermittent(n=220)

Identified gram-negatives 29 (72.5%) 31 (72.1%)

Pulmonary Infection 115 (54.2%) 120 (54.5%)

APACHE II 21 (17-26) 21 (16-25)

Patient CharacteristicsBLISS• Same total daily dose of antibiotics regardless of infusion

strategy

4/3/2017 53Abdul-Aziz, MH, et al. Intensive Care Med 2016;42:1535-1545

Group characteristics Continuous(n=70)

Intermittent(n=70)

P. aeruginosa 37% 31%

A. baumannii 25% 23%

Pulmonary Infection 46 (66%) 36 (51%)

Concomitant antibiotic 33 (47%) 33 (47%)

APACHE II 21 (17-26) 21 (15-26)

Clinical OutcomesBLING II & BLISSStudy BLING II (2015) BLISS (2016)

Results - No statistically significant difference between groups in any outcome

- Clinical cure at 14 days higher after cessation of antibiotic

- 56% vs. 34%, p=0.011

Limitations - 26% of patients had renal replacement therapy

- Median length of treatment: 3 days

- Median length of treatment: 7 days


Extended Infusion Carbapenems?• Limited data available – mainly case reports

• Risk documented with extended infusion doripenem– Doripenem 1g q8h over 4 hours for 7 days vs.

imipenem-cilastatin 1g q8h over 30 minutes for 10 days– Study stopped early due to inferior efficacy and

increased mortality– Limitation: first dose was the extended infusion

4/3/2017 55Kollef et al. Critical Care 2012, 16:R218Osthoff M, et al. Swiss Med Wkly. 2016;146:w14368

Indications• Confirmed P. aeruginosa infections

• Critically ill patients

• Pulmonary infections

• Severe sepsis

• Institution-specific resistance patterns


Dosing Strategy• Loading dose before using extended infusion or

continuous infusion

• Quicker attainment of MIC with loading doses

• Start extended infusion dosing schedule at next interval

4/3/2017 57

Other Hospitals that Use Extended or Continuous Infusions• Out of 17 responses:

– All hospitals used extended or continuous infusion beta-lactam therapy

– All hospitals used piperacillin-tazobactam as an extended infusion

– Five hospitals used meropenem extended infusion

– Three hospitals used cefepime extended infusion

4/3/2017 58

Other Hospitals that Use Extended or Continuous Infusions• University of Missouri Health Care• Oklahoma Heart Hospital• University of California – Davis Medical Center• UC – San Diego Health• Detroit Medical Center• University of Chicago• Beth Israel Deaconess Medical Center• Gulf Coast Regional Medical Center• UW Health• UCLA• Johns Hopkins• University of Utah

4/3/2017 59

Obstacles for Extended Infusion Beta-lactam Implementation

4/3/2017 60

IV Medication Compatibility• Doses can run for 12 hours each day• Requires multiple peripheral lines or a central line

4/3/2017 61

Piperacillin-tazobactam

Cefepime Meropenem

- Levofloxacin, vancomycin

- Diltiazem, dobutamine, hydralazine, labetalol, nicardipine

- Famotidine, prochlorperazine, promethazine, pantoprazole, insulin

- Midazolam

- Diltiazem, dobutamine, nicardipine

- Diphenhydramine, propofol

- Famotidine, pantoprazole, prochlorperazine, promethazine

- Midazolam, dopamine, morphine

- Magnesium sulfate

- Amiodarone,nicardipine

- Diazepam, propofol- Pantoprazole

*Not an inclusive list of incompatibilitiesTrissel’s 2 Database

Cost Benefits of Extended Infusion Beta-lactams• Reduced cost of piperacillin-tazobactam

– Decrease of one dose per day– Potential decrease in length of hospital stay

• Possible decreased length of ICU stay and hospital stay with other beta-lactams– No decrease in total daily drug dosage

4/3/2017 62Brunetti L, et al. Ann Pharmacother 2015; 49:754Schmees PM, et al. Am J Health Syst Pharm 2016; 73:S100

Review Question 4• Which of the following is an area where

pharmacists can make an impact on proper use of extended infusion beta-lactams?A. Administration technique of extended infusion beta-

lactams

B. IV incompatibility surveillance

C. Diagnosis of an infectious disease process

D. All the above

4/3/2017 63

Steps Toward Implementation

4/3/2017 64

Nursing• Education on benefit and purpose of extended

infusion therapy

• Maintain cleanliness of line sites to prevent infections

• PT / OT issues with IV administration of medications

4/3/2017 65

Pharmacy• Configuration of smart pumps and CPOE entries

• Selection of appropriate patient

• Considerations: – IV incompatibilities– Line access

4/3/2017 66

Summary• Time-dependent properties allows extended-infusions

• Piperacillin-tazobactam has the most evidence for use

• Extended infusion beta-lactams should be used in:– Critically ill patients– Pulmonary infections

• Patient outcomes are improved when used appropriately

• Multidisciplinary implementation is required for successful extended-infusion beta lactams

4/3/2017 67

Acknowledgement• Kane Hosmer, PharmD, BCOP

4/3/2017 68

Combating Antimicrobial Resistance with Extended Infusion Beta-lactamsStephen Andrews, PharmDPGY-1 Pharmacy Practice [email protected]

Documents

Combating Antimicrobial Resistance with Extended Beta ... · PDF fileObjectives • Review pharmacokinetic and pharmacodynamic (PK/PD) properties of beta-lactam antibiotics • Evaluate